Affiliation:
1. SASTRA University, India
2. Alagappa University, India
3. Compunnel, USA
Abstract
Higher resolution images are integral across diverse applications due to several compelling reasons. Firstly, they offer superior detail and clarity, making them indispensable in fields such as medical imaging, satellite observations, and scientific research where capturing intricate details is paramount. In medical imaging, high resolution is pivotal. Despite the advantages of high-resolution images, they are not always accessible due to the costly setup required for high-resolution imaging. Feasibility may be constrained by essential limitations in sensor optics manufacturing technology. To overcome these challenges, cost-effective deep learning methods can be employed. In this context, the proposed holistic transformer super-resolution technique aims to enhance the resolution of an image beyond its original level.
Reference12 articles.
1. Deep reparametrization of multi-frame super-resolution and denoising.;G.Bhat;Proceedings of the IEEE/CVF International Conference on Computer Vision,2021
2. Real-world single image super-resolution: A brief review.;H.Chen;Information Fusion,2022
3. Image super-resolution using deep convolutional networks.;C.Dong;IEEE Transactions on Pattern Analysis and Machine Intelligence,2015
4. Image interpolation techniques in digital image processing: An overview.;S.Fadnavis;International Journal of Engineering Research and Applications,2014
5. Comparisons between different interpolation techniques.;G.Garnero;The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences,2014